关键词: 光伏组件, 噪声特性, BP神经网络, 小波神经网络, 粒子群算法, 遗传算法

Abstract: In order to monitor the working state of each PV module, which can be realized by the carrier communication technology that uses the photovoltaic series as media, it is necessary to master the channel noise characteristics of photovoltaic series. In this paper, based on the measured photovoltaic channel noise in a photovoltaic power plant, a particle swarm optimization BP neural network is proposed to model the channel noise in the photovoltaic serial channel. The experimental and simulation results show that the predicted output of the particle swarm optimization BP neural network model have a consistent variation trend with the tested original noise in power spectral density and time domain waveform, which proves the effectiveness of the proposed model. Compared with the wavelet neural network and BP neural network optimized by genetic algorithm, the BP neural network with particle swarm optimization has less RMS error and higher accuracy.

Key words: photovoltaic module, noise characteristics, BP neural network, wavelet neural network, particle swarm optimization, genetic algorithm